A Model for Route Learning in Opportunistic Networks
Jorge Visca, Javier Baliosián · NOMS 2022-2022 IEEE/IFIP Network Operations and Management Symposium · 2022
Opportunistic Networks (ONs) are complex systems where nodes move and meet stochastically. For data to be trans-mitted, it must be copied between nodes and then carried around through multiple hops. Because the network is continuously evolving and can be highly fragmented, the usual connectivity graph representation, typically used to compute routes or doing offline traffic engineering, is not suitable.This article proposes a model specially designed for route management on ONs, and we show its usefulness by implementing an ML-based route management algorithm for those networks. This algorithm is trained offline over the model built from historical traces of a particular network setup and then deployed to make fast, real-time forwarding decisions.